memcg: shrink obj_stock_pcp and cache multiple objcgs
Status from this message: [merged mm-stable] memcg-multi-objcg-charge-support.patch removed from -mm tree
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v3 Latest4 patches 1 replyApplied 4/4
- v24 patches 11 repliesSuperseded
- v14 patches 13 repliesSuperseded
Changes in v3 from the cover letter
http://lore.kernel.org/[email protected]
- Fix comments (Muchun & Qi)
- Simplify code (David Laight)
- Fix handling of archs with base page size larger than 256 KiB (Sashiko)
Patches in v3 All 4 applied
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/4 | memcg: store node_id instead of pglist_data pointermm/ | +19 | 0 replies | Applied |
| 2/4 | memcg: uint16_t for nr_bytes in obj_stock_pcpmm/ | +19 | 0 replies | Applied |
| 3/4 | memcg: int16_t for cached slab statsmm/ | +12 | 0 replies | Applied |
| 4/4 | memcg: multi objcg charge supportmm/ | +142 | 1 reply | Applied |
Reviews trailers per patch
All 4 patches have a Reviewed-by or Acked-by.
| Patch | kernel test robot | Muchun Song | Harry Yoo (Oracle) | Qi Zheng |
|---|---|---|---|---|
| Series | T-b | |||
| 1/4 | T-b | A-b | R-b | A-b |
| 2/4 | T-b | A-b | R-b | A-b |
| 3/4 | T-b | A-b | R-b | A-b |
| 4/4 | T-b | R-b |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 4 reviewers.
Cover letter [PATCH v3 0/4] memcg: shrink obj_stock_pcp and cache multiple objcgs
Commit 01b9da291c49 ("mm: memcontrol: convert objcg to be per-memcg per-node type") split a memcg's single obj_cgroup into one per NUMA node so that reparenting LRU folios can take per-node lru locks. As a side effect, the per-CPU obj_stock_pcp -- which caches a single cached_objcg pointer -- thrashes on workloads where threads of the same memcg run on different NUMA nodes. The kernel test robot reported a 67.7% regression on stress-ng.switch.ops_per_sec from this pattern.
Commit d0211878ce06 ("memcg: cache obj_stock by memcg, not by objcg pointer") landed as a temporary fix by treating sibling per-node objcgs as equivalent for the cache lookup, intended to be reverted once per-node kmem accounting is introduced. This series takes a more general approach: cache multiple objcgs per CPU using the multi-slot pattern memcg_stock_pcp already uses, so the per-node objcg variants of one memcg can all coexist in the stock without ever forcing a drain. The temporary fix can then be reverted.
To avoid increasing the per-CPU cache footprint, the first three patches shrink the existing single-slot obj_stock_pcp fields. The final patch converts cached_objcg and nr_bytes into NR_OBJ_STOCK=5 slot arrays and reorders the struct so the entire consume/refill/account hot path fits within a single 64-byte cache line on non-debug 64-bit builds (verified with pahole).
Shakeel Butt (4):
memcg: store node_id instead of pglist_data pointer
memcg: uint16_t for nr_bytes in obj_stock_pcp
memcg: int16_t for cached slab stats
memcg: multi objcg charge support
mm/memcontrol.c | 214 +++++++++++++++++++++++++++++++++++------------- 1 file changed, 157 insertions(+), 57 deletions(-)
--
Changes since v2:
http://lore.kernel.org/[email protected]
- Fix comments (Muchun & Qi)
- Simplify code (David Laight)
- Fix handling of archs with base page size larger than 256 KiB (Sashiko)
Changes since v1:
http://lore.kernel.org/[email protected]
- Collected review tags (Harry & Muchun)
- Fix comparison operators (Harry)
- Use round robin for drain
2.53.0-Meta